MLB Replay Review

Now that we’ve had a good half season or so of replay review in baseball, I’m curious what people think of it. Personally, I’m mostly happy with how it’s been working, although there are obviously still some issues to hash out.

For instance, I’m surprised at how often plays do not get overturned, despite what looks like fairly convincing evidence. Maybe the standards of evidence are still a little too high. To me, it seems more like a ‘preponderance of evidence’ than a ‘beyond a reasonable doubt’ kind of situation. I also don’t see why replay officials should be biased toward the original call on the field. Let the people with the best evidence make the call.

The most controversial calls have involved plays at the plate. Recently, on my fantasy baseball site, there was discussion of a key play during the July 31 game between the Reds and Marlins. The Reds were down 1-0 in the 8th, with the bases loaded and 1 out. Frazier hit a shallow fly ball to right field, and Cozart tagged up at third. The ball beat him at home by quite a bit, and Cozart ran awkwardly by the plate, getting tagged without sliding. The Marlins left the field, thinking the double-play had ended the inning. But Reds manager Price requested a review.

What happened next was remarkable. The crew in New York spent over six minutes analyzing the play, making this the longest review ever. Here’s the video:

You can see the play right away, but, if you have time, the analysis / time-killing by the Reds broadcasters is kind of interesting. In the end, the call on the field was overturned. Cozart got credit for the run, the Marlins manager was kicked out of the game, and two more Reds scored on the next at bat. The Marlins, playing under protest, ended up losing 3-1.

The reaction to this review was overwhelmingly negative. Here’s Marlins manager Mike Redmond after the game:

As a former catcher in this league for 13 seasons, as a grinder, as a guy who loves this game and respects this game so much, this game has been a part of my life forever, but to lose a ballgame tonight on that play is a joke. It’s an absolute joke. I don’t think anybody who plays this game should feel good about winning that game. And I would say that if [the situation] had been reversed. That guy was out by 15 feet. It was a great baseball play.”

(I suspect the language has been cleaned up a little, especially around the words ‘joke’.)

And here’s some fairly typical analysis from Deadspin’s Sean Newell:

The rule is a mess. Even after the learning experience that was the debacle with Russell Martin in June, no one knows what the hell they should be doing as a play unfolds on the field. […] It’s clear that Mathis is standing in front of the plate, but…where is he supposed to stand? And we don’t really know what “in order to field a throw” means. When is a catcher deemed to have started to field a ball? Is anticipating where the ball will land fielding it? It looks awful, too, because Cozart is clearly dead-to-rights. But confusion reigns: he is unsure of what to do and doesn’t slide, so after the play you can see him raise his arms as if to say “WTF, man?” There was no collision and no contact except for the tag. The purpose is obviously good—no one wants to see Buster Posey get blown up again—but it’s proving to be near impossible to negotiate this rule in-game to any degree of satisfaction.

The key to the decision is the controversial Rule 7.13, which states:

(1) A runner attempting to score may not deviate from his direct pathway to the plate in order to initiate contact with the catcher (or other player covering home plate). If, in the judgment of the umpire, a runner attempting to score initiates contact with the catcher (or other player covering home plate) in such a manner, the umpire shall declare the runner out (even if the player covering home plate loses possession of the ball). […]
Rule 7.13(1) Comment: The failure by the runner to make an effort to touch the plate, the runner’s lowering of the shoulder, or the runner’s pushing through with his hands, elbows or arms, would support a determination that the runner deviated from the pathway in order to initiate contact with the catcher in violation of Rule 7.13. If the runner slides into the plate in an appropriate manner, he shall not be adjudged to have violated Rule 7.13. […]
(2) Unless the catcher is in possession of the ball, the catcher cannot block the pathway of the runner as he is attempting to score. If, in the judgment of the umpire, the catcher without possession of the ball blocks the pathway of the runner, the umpire shall call or signal the runner safe. Notwithstanding the above, it shall not be considered a violation of this Rule 7.13 if the catcher blocks the pathway of the runner in order to field a throw, and the umpire determines that the catcher could not have fielded the ball without blocking the pathway of the runner and that contact with the runner was unavoidable.

Interestingly, the powers that be were clearly aware that this rule might be controversial, as they included this pre-amble: “The Playing Rules Committee has adopted Rule 7.13 as set forth below on an experimental basis for the 2014 season.”

Sure enough, they got their controversy!

That said, I agree with the decision in this case–and not just because it led to the Reds winning the game! Admittedly, the optics are quite bad. Cozart was out by a mile, and doesn’t even try to slide (or touch the plate). Still, this is exactly the sort of play that used to lead to a big collision at the plate. So, in that sense, it worked exactly as it was supposed to. If the MLB is going to ban collisions at the plate (as they should), then they have to strictly ban catchers from blocking the plate. It’s two sides of the same coin.

Still, there are some problems with how the rule is currently set up. First, catchers have the habit of blocking the plate in these situations, and it’s hard not to do that in the heat of the moment. This problem will take care of itself with a bit of time, as catchers retrain their instincts on these plays.

A bigger problem concerns when to start calling interference. If the runner was halfway between third and home when the catcher had the ball in front of the plate, then we can surely agree that this should not count as interference. Likewise, if the runner arrives right after the catcher catches the ball blocking the plate, then we can all agree that it is interference. But it’s hard to see where you draw the line between these two extremes. At what point does interference begin? In the Cozart play, I think a case can be made for interference. Cozart could have decided to barrel into the catcher, hoping to knock the ball loose. There’s a long history of this happening in baseball, even when the catcher has the ball well in advance. Here’s a particularly remarkable example:

So, if there was potential for a collision in the case of Cozart, which there was, then you have to enforce the corresponding interference rule. The reviewers took way too long, but they got the play right.

Looking forward, what we need is a clear way of showing exactly when interference happens. Here’s my proposal. Put a line on the field, roughly 25 feet up the third base line from home. If catcher receives the ball when the runner is past this line, then he can’t block the plate. Otherwise, he can. (It’s sort of like the circle they use in basketball to determine whether to call a charge.) This would give the umpires a clear (albeit somewhat arbitrary) way of making the call. It would give the catchers something to look at in deciding how to approach the play, and it would be minimally disruptive of the existing game.

I believe this is the best solution to an inevitably awkward situation.

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

13 Responses to MLB Replay Review

  1. Jon says:

    Interesting piece…I wonder how this rule is applied if a runner gets into a rundown between home and third. As it stands now, it seems the catcher would be blocking the plate anywhere along the baseline. Your proposal could take care of some of that.

  2. David says:

    Now I know why Toronto fans love Buck Martinez!

    As for replay, I’m generally ok with it, too, but I really don’t like the current managerial practice of moseying out to the field and killing time until a thumbs up or thumbs down from the dugout. That needs to be addressed before next season.

    As for the problems at home plate, it seems to me that the real root of the problem is that the catcher has to tag the runner. This is what led to collisions back in the day and now leads to confusion. If it were up to me, I would rewrite the rules so that there is always a force out at home plate–just like first base. Catchers wouldn’t have to worry anymore about tagging runners, and all these problems would be solved.

  3. Nates says:

    OK, since overthinking stuff is what this blog is all about, let’s dive further in. Here are some criteria for a successful solution to the problem:
    (1) Prevents (or at least greatly reduces) home plate collisions
    (2) Easy to assess
    (3) Maximizes fun and excitement
    (4) Minimally disrupts baseball tradition
    (5) Intuitive / Non-arbitrary

    Some of these criteria may be in tension with one another, but I hope we can agree that collectively maximizing them is a good thing. (1), (2) and (3) should be self-evident. The thought behind (4) is something like this: beware of unintended consequences. Baseball has been served well by a tradition of careful, incremental changes to the game. The thought behind (5) is that the rule should just make sense, given the broader principles and goals of baseball.

    So, we have two proposals: David’s (force at home on every play) and mine (catcher can block plate when runner is beyond 25 foot line). Both satisfy criteria (1) and (2). Mine potentially has issues with (5). After all, why 25 feet, exactly? It’s a little arbitrary, admittedly, but I would argue that it roughly captures when the threat of a collision play would begin. Maybe.

    But with David’s proposal, I’m still a little worried about criteria (3) and (4). On maximizing fun, it would be a shame to lose all the great sweeping tag plays at home. Stuff like this:

    Or, even better, this:

    And then there’s criteria (4): disruptiveness. I could see David’s proposal having some undesirable consequences. For example, would it make for some unnatural looking home-to-first double plays? And what happens if the runner turns around and heads back to third? (Actually this might be more of an issue with criterion (2): how do we determine when the force play kicks in? What if the runner takes a running lead, but then heads back on the grounder? When does the commitment to running home begin?

    I still lean toward my proposal, but I’m willing to hear more.

    • David says:

      I like these criteria.

      Yes, my proposal involves sacrificing some fun and excitement for the sake of safety, as does yours. But you suggest that there is one exciting element of the game that your proposal, but not mine, saves–viz., the sweeping tag! The jump-over-the-catcher! What can I say? For the sake of safety and fewer confused disputes, I’m willing to let these plays go the way of the fun and exciting steamroll-the-catcher move, which we both agree was appropriately retired. And it’s not like there wouldn’t be any more excitement at home–we’d still have runners trying to beat the ball to the plate in situations where the run will decide the game.

      You write: “I could see David’s proposal having some undesirable consequences. For example, would it make for some unnatural looking home-to-first double plays? And what happens if the runner turns around and heads back to third? (Actually this might be more of an issue with criterion (2): how do we determine when the force play kicks in? What if the runner takes a running lead, but then heads back on the grounder? When does the commitment to running home begin?”

      Well that’s easy. Just put a line on the field, roughly 25 feet up the third base line from home. Once the runner crosses that line it becomes a force out at home. 😉

  4. Nates says:

    Hmm, we seem to be converging on the same point…

    With both proposals, we get no collisions, an easy to assess rule, and a goofy new line intersecting the third base path to home. But my proposal has the bonus of exciting tag plays at home. Isn’t that the tie breaker?

    Also, where’s Josh in this debate? I know his summer vacation is over, but I’m pretty sure he has a rant against the whole replay system building up inside…

  5. Josh says:

    Okay I’ll throw in my two cents. I don’t think I agree with the preference for safety both of you seem to agree upon. I’ve thought similar things about the “neighborhood play” issue – like there’s some sort of right for the defensive team to be able to turn a double play they aren’t actually able to turn. If safety is the concern, can’t this be solved by players not doing things that will cause them injury? The Pete Rose example always comes up here, but that seems more to do with its having been during an all-star game, when the catcher perhaps did not expect it. But the catcher is wearing armor. If the worry is about the runner being injured, can’t they just not slide in a way that is likely to get them injured?

    (Also this has moved a bit away from instant replay, which is fine… though it’s interesting to me that instant replay has shined some light onto rules 0 like this one – that had for a time been ignored because they couldn’t be enforced in such a way in real time).

    One issue I’d like to raise about replay (akin to David’s frustration with managers sauntering out onto the field) there does seem to me to be some tension between getting the call right and it decreasing the entertainment value of the game. Aren’t there times when we’d prefer the dramatic finality on the field to the later final determination of truth?

    A hypothetical situation – the world series rides on a play at the plate (notwithstanding proposed rules changes). The umpire calls the runner safe, the crowd storms the field 70’s style as the home team has just won the game. That scene can no longer happen today as a challenge is more or less assumed to need to take place. So instead of the excitement of the on-the-field call, we now have 3-5 minutes of waiting around and a sort of suppressed cheer if the final call is upheld. Insofar as baseball is a form of entertainment, it seems to me at least worth asking whether we might not prefer the drama to a sort of precision that takes away from that drama. Any thoughts?

  6. Nates says:

    Well runners do get hurt, but it’s the catchers that David and I (and others) are worried about. Despite all the armor they’re wearing, the law of momentum is working against them: p = mv
    The runner has all the velocity, meaning the catcher takes most of the impact. (Just ask Buster Posey.)

    Yes, players could just change how they play, but what we’re identifying is a perverse incentive that encourages injury-risking behavior. Bashing into the catcher is an effective strategy when you arrive after the ball. So it’ll always be tempting if it’s a legal option.

    (On the neighborhood plays at second base, my sense is that the injuries here could mostly be prevented if the umps called runners out for slides that miss or go past the bag. That’s already in the rules, so it just needs to be enforced.)

    On replay, yes there will be occasional situations where replay will rob us of a dramatic ending. There have also been situations where the lack of replay left us feeling frustrated and robbed of something great: like Galarraga’s almost-perfect game:

    I don’t know how you quantify the potential losses or gains, but my hunch is that they roughly cancel each other out. Given the added epistemic virtue of getting the call right, my vote is for replay. (Especially if they can work on speeding it up a little.)

  7. Josh says:

    Regarding the injuries – in the case of the neighborhood play, there is a rule about sliding but a lot of the time – seemingly almost every time – the runner has acted according to the rule, as all they have to do is touch the base with some part of their body on the way by.

    This is relevant to my point about the value of truth – I think (could be wrong) MLB explicitly ruled out the use of replay for the neighborhood play. One way to understand this would be they decided that knowing the truth was less important than preserving the drama of the double-play (even though some of those double plays aren’t really double plays because the fielder didn’t actually touch the base while in possession of the ball). If what I’ve said is true, this is a situation where the powers that be (perhaps wrongly) don’t think truth is as important as something else. My own sense is this has to do with denial abetted by history – we’ve gotten used to double plays that aren’t double plays, and we don’t want that taken from us. Now I’m not saying that’s it’s right to do that, I’m just pointing out that the truth isn’t the only or even the dominant value at work here in other situations. So our baseball “intuitions” are not all truth-centered.

    Regarding non-double-play plays at home (and other bases) – something similar had been happening prior to replay – “ball beats runner = out” was quite useful as a compromise, and seemed to minimize arguments and satisfy fans. Replay has taken that away, and that might be a bad thing from the perspective of what many fans are used to.

    This brings me to another point perhaps worthy of discussion: many of the current rules were designed with a different level of epistemic access to truth in mind, right? Many rules would have been written differently if their ability to enforced via replay had been assumed during that writing. One example I’ve noticed of this is caught-stealing. It seems (obviously totally anecdotally) that the runner sliding into second is safe way more often than what we’re used to, especially upon review.

    If this is true, there are two possible replies: (1) replay is helping us get the call on stolen bases right, and if that means more stolen bases, so be it. (2) stolen bases have historically been a harder thing to acquire, and we’d like to maintain that difficulty, both to preserve the sanctity of records and for competitive balance. So perhaps we need to change the stolen base rule to, for example, make it a force out (like David’s proposal at home plate) or something similar.

    I’m not saying (2) is right in this instance, but I’m saying something LIKE (2) seems likely to be right in at least some instances, and we need to be mindful of the ways new replay options will affect the enforcement of rules that has been hitherto different.

    Regarding speediness of replay: I think this is the biggest no-brainer here. The current system of “challenges,” borrowed I guess from football and tennis, is unbelievably stupid. It creates this manager-centered pseudo-drama (the sideline/dugout closeups, the headset and dugout phone conversations, the slow strolls out onto the field to buy time, etc.) that is totally avoidable.

    Here is my proposal – call it “passive replay.” Every game should have a 5th umpire (either on site next to tv monitors or in this New York office) – perhaps even every 2 games, if it’s too hard to retain 15 of these people. That 5th umpire is in charge of watching for close plays. Whenever a play is close enough to be reviewable, that 5th umpire contacts the on-the-field crew chief via ear-bud and lets them know things need to be delayed a bit. This could happen under the radar much of the time, so the fans wouldn’t even have to know. They’d be like “so I know that throw looked close but the first replay showed obviously you guys got it right (or wrong), so just keep playing.”

    Some of the time when it was much closer and more ambiguous, the 5th umpire would deliver some sort of message to that effect and then the crew chief would have a recognizable sign that the play was being reviewed. Like the current “umpire review.” Then they’d watch it and decide whether to overturn in.

    On this system, there’s no need for team-originated challenges. If it’s close enough to look, the umpire crew will take care of it. If it’s not, they won’t.

    This also addresses the concern about “preponderance of the evidence” vs. “conclusive proof.” Since it’s not a coach challenging, there are no real burden-of-proof issues. The umpires (5th included) are deciding among themselves what happened, and so their internal reasoning becomes less a matter of adversarial challenges used/”wasted”/whatever, and more just an issue of getting it right.

    One more idea – don’t we need an equivalent of the NFL “whistle” rule? I was surprised to see a play in which, a runner on first, a batted ball was judged trapped rather than caught by a left fielder (he caught the ball in shallow left-center, easily controlled it and came up throwing for first, attempting a double play on the premise that he had caught it – the runner was midway between 1st and 2nd). When the play was overturned, the umpires ended up placing runners in 1st and 2nd, when for sure, had the call been made as “trap” initially, the runner at 2nd would have been forced out by a mile. Since the call on the field sent the runner back to 1st, and since there was no force available at second, the outfielder didn’t throw the ball to second, but instead on to first in an attempt at a double play. When the call was overturned, the batter was given first and the runner second. The point is that the call on the field altered the fielder’s choices, and in a way that totally make sense. Do the umpires have the discretion to call that runner out on the grounds that had the original call been made differently, that’s what would have happened? And if they don’t, why don’t they? The NFL whistle rule acknowledges that calls sometimes change player expectations in an irreversible way. Something like that needs to happen in baseball too.

  8. Josh says:

    Also regarding catches – if the catcher already has the ball, can’t they sidestep the runner and make the tag? If they don’t already have the ball, can’t they just get out of the way as they weren’t getting an out anyway? And if the ball’s just arriving, can’t we just make a rule that requires the runner to slide and not collide with the catcher unnecessarily? My little league had such a rule and it worked fine most of the time. If the danger is towards the catcher, can’t we just have something like an “unnecessary roughness” call that can be made towards the runner, instead of giving much easier force-outs or creating a line that dictates when the plate may be blocked?

    (btw this rule is a good example of a rule that’s come under more scrutiny just because replay challenging of it is not possible, and perhaps was written with different epistemic standards in mind)

  9. Nates says:

    OK, we have a lot of different points in play here. Let me try to address a few of them.

    On neighborhood plays, I’m pretty sure it was risk-of-injury rather than drama-preservation that was the main reason for not having replay in these cases. (At least, that’s my recollection of the debate. See here for supporting evidence.)

    Also, it’s not quite right to say that a slide is legal if the runner touches the base. That is how it’s called in practice, but look at the actual rule 7.08 b:

    Any runner is out when […]
    (b) He intentionally interferes with a thrown ball; or hinders a fielder attempting to make a play on a batted ball;
    Rule 7.08(b) Comment: A runner who is adjudged to have hindered a fielder who is attempting to make a play on a batted ball is out whether it was intentional or not. If, however, the runner has contact with a legally occupied base when he hinders the fielder, he shall not be called out unless, in the umpire’s judgement, such hindrance, whether it occurs on fair or foul territory, is intentional. If the umpire declares the hindrance intentional, the following penalty shall apply: With less than two out, the umpire shall declare both the runner and batter out. With two out, the umpire shall declare the batter out.

    Notice: touching the base only gets the runner off the hook if there’s no intentional interference. I’m suggesting that we should enforce intentional takeout slides–even the ones where the runner touches the base. You know, like this one:
    Holliday takes out Scutaro

    On stolen bases, the overall average success rate so far in 2014 is 73%. In 2013, it was 72%. 2012: 73%. Not much has changed. To be honest, I’m surprised by this, as I would’ve expected more of an impact from replay. Apparently the bad calls before replay were evenly distributed between safe and out.

    But I do concede your broader point, which is that many baseball rules and practices were not designed for this level of scrutiny. Of course, they weren’t designed for 6’10” pitchers either, but we’re mostly making do. I suspect that various tweaks will be required over the coming years to establish a new effective equilibrium that includes replay.

    I like the passive replay idea. It wouldn’t surprise me if that’s where we’re headed. But this, in itself, wouldn’t speed up the process. (Replay officials aren’t currently waiting for managers to make a challenge: they’re preemptively checking close plays, anticipating possible challenges. So we need to come up with a way of speeding up the review process itself. I think this requires more eyeballs, not less. So, it’s probably better off in the hands of a dedicated video-center staff, rather than being left up to a single 5th-umpire on site. Still, I agree that they should be able to overturn calls whenever they see evidence for it.

    And, finally, back to catchers. An unnecessary roughness rule would probably do the trick. My thought was that a line across the base path would have the advantage of being less subjective to call, but, otherwise, both proposals accomplish what we want here.

  10. Josh says:

    So my point about the neighborhood play is that to some extent, “avoiding injury” as goal presupposes that the play causing the injury is inevitable or somehow desirable. If we believe a play “should” happen, we’re more likely to shield it from the prying eyes of instant replay. If there were a deterrent in place like a fielder knowing that if they don’t actually touch the base while holding the ball, the out won’t be called, then this combined with the risk of injury might deter them from engaging in the behavior. We don’t currently have rules against doing a whole host of other dangerous things, like intentionally running into the backstop, because there’s no reward for them either. If the sense of entitlement to an out weren’t there, don’t you think fewer people would attempt the dangerous maneuver in the first place?

    I agree that baseball wasn’t designed for 6’10” pitchers, but that’s categorically different than the epistemic conditions of umpiring changing don’t you think? Continued enhancements in physical fitness definitely have changed some expectations about play, but in a different way than being able to discern what actually happened on a given play from the standpoint of officiating might have.

    I was thinking about the passive-replay idea – I do think it would enhance speed only because there wouldn’t be that pause involving the manager. I’m sure you’re right that they’re already looking into things, but if it just wasn’t even an option, all those times when the challenge ends up not happening wouldn’t result in the stroll-out-stroll-back of the manager.

    Broader point – I’ve never understood why argument of any sort is tolerated during a game. It wastes time and absent an actual rules challenge (which basically never happens) they’re just jabbering on about judgment calls that won’t be reversed. I’d be totally okay with a zero-tolerance rule for arguments.

    Another virtue of the passive replay idea is that it’s sort of strange we’re placing managers in charge of flagging bad calls. What if they run out of challenges? It’s not their fault that the 4th call the umpire makes is bad, even if the first four seemed good. Right now we make it a somewhat risky (admittedly not that risky, but still risky) move for the manager to make a challenge.

    If money were no obstacle, yeah, having lots of people watching would be good. “5th umpire” is I guess figurative – there could be a team, an individual, they could be on-site, or not.

    Reflecting on the catcher issue – it occurs to me the NFL went through something similar back in the late 80’s/early 90’s with the “in the grasp” rule. The idea was to prevent quarterback injuries by whistling the play dead when the quarterback was “in the grasp.” This led to a bunch of plays where it seemed like the quarterback might yet throw the ball or escape a tackle being ended, and the game suffering as a result. I think they stopped calling this so much in the seasons that follow. There is still, of course, a “roughing the passer” penalty.

    So the same thing might be happening here – better to disincentivize the bad behavior vis-a-vis the player who is doing it (crashing into the catcher, body-slamming a quarterback) than make a more paternalistic rule that tries to prevent the situation from arising. I believe this might be broadly in line with t-lib (which I’m totally okay with as regards sports).

    Also perhaps evidence that truth is just not valued as much as people seem to think – why not have a computer call balls and strikes? This seems to be totally out of the question but of course it’d be more accurate than letting the umpire set “their” strike zone…

  11. Nates says:

    Yeah, that all sounds pretty reasonable. It’s funny, I find myself feeling an odd fondness for the crusty old manager coming out to yell at the ump, kick some dirt around, etc. I agree that there’s no epistemic value to it. I think it’s just old-timey nostalgia. I suppose as time-wasters go in baseball, it’s at least entertaining (when done well).

    It seems like there’s an emerging consensus that baseball needs to improve its pacing. Dramatic delays aren’t so bad, but the batters and pitchers walking all over the place between every pitch is painful, totally unnecessary and easily fixed. Also, visits to the mound. If the pitcher’s in trouble, he should either fight his way out of it or get pulled.

  12. Josh says:

    I know what you mean about the nostalgia for manager arguments… I was at the White Sox’s tony LaRussa appreciation day last week, and they kept showing the video of him uprooting 3rd base. there is a sort of catharsis for the fans in watching the futile arguments of the managers. Sadly replay changes it into this time-wasting non-discussion that doesn’t even seem like an argument, since it displaces the locus of disagreement towards some nebulous point out in the ether halfway between the game and the “New York office.”

Leave a Reply

Your email address will not be published. Required fields are marked *